Single Cell SCLC Cell Line Archetypes

The purpose of the notebook is to apply PCHA to the single cell cell line data.

Read single cell data

Reading in preprocessed (MAGIC imputed) data from from 2-Human_cell_lines_scPCHA.ipynb

# file = "../../out/cell-lines/X_magic_for_R.h5ad"
# Convert(file, dest = "h5seurat", overwrite = TRUE)
# data <- LoadH5Seurat("../../out/cell-lines/X_magic_for_R.h5seurat")
# Idents(data) = data@meta.data$cline

ElbowPlot(data) #this plot indicates  the top 8 are important

DimPlot(data, reduction = "pca", label = TRUE)

pct <- data[["pca"]]@stdev / sum(data[["pca"]]@stdev) * 100

# Calculate cumulative percents for each PC
cumu <- cumsum(pct)

# Determine which PC exhibits cumulative percent greater than 90% and % variation associated with the PC is less than 5
co1 <- which(cumu > 90 & pct < 5)[1]

print(co1) #38

[1] 38

# Determine the difference between variation of PC and subsequent PC
co2 <- sort(which((pct[1:length(pct) - 1] - pct[2:length(pct)]) > 0.1), decreasing = T)[1] + 1

# last point where change of % of variation is more than 0.1%.
print(co2) #11

[1] 11

data <- FindNeighbors(data, dims = 1:10)

Computing nearest neighbor graph
Computing SNN

data <- FindClusters(data, resolution = 0.5)

Modularity Optimizer version 1.3.0 by Ludo Waltman and Nees Jan van Eck

Number of nodes: 16108
Number of edges: 432634

Running Louvain algorithm...

0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|

Maximum modularity in 10 random starts: 0.9446
Number of communities: 21
Elapsed time: 1 seconds

data <- RunUMAP(data, dims = 1:10)

10:49:36 UMAP embedding parameters a = 0.9922 b = 1.112
10:49:36 Read 16108 rows and found 10 numeric columns
10:49:36 Using Annoy for neighbor search, n_neighbors = 30
10:49:36 Building Annoy index with metric = cosine, n_trees = 50
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
10:49:38 Writing NN index file to temp file /var/folders/vh/xk34gq593k53lzm7wlyg64xw0000gn/T//RtmpIr3N9b/file125ad65ea1dbd
10:49:38 Searching Annoy index using 1 thread, search_k = 3000
10:49:41 Annoy recall = 100%
10:49:42 Commencing smooth kNN distance calibration using 1 thread
10:49:43 Initializing from normalized Laplacian + noise
10:49:46 Commencing optimization for 200 epochs, with 571528 positive edges
0%   10   20   30   40   50   60   70   80   90   100%
[----|----|----|----|----|----|----|----|----|----|
**************************************************|
10:49:54 Optimization finished

DimPlot(data, reduction = "umap")

Plot important SCLC Genes and add cell line labels

DimPlot(data, label = TRUE) 
VlnPlot(data, features = c("ASCL1", "YAP1", "POU2F3", 'CALCA','NEUROD2', 'MYC'), combine = FALSE)

[[1]]

[[2]]

[[3]]

[[4]]

[[5]]

[[6]]

Plotting the data

library(dplyr)
data.markers <- FindAllMarkers(data, only.pos = TRUE, min.pct = 0.25, logfc.threshold = 0.25)
data.markers.top <- data.markers %>% group_by(cluster) %>% top_n(n = 30, wt = avg_log2FC)

# install.packages('plotly')

# Load plot_ly
library(plotly)

# Extract tSNE information from Seurat Object
pca_1 <- data[["pca"]]@cell.embeddings[,1]
pca_2 <- data[["pca"]]@cell.embeddings[,2]
pca_3 <- data[["pca"]]@cell.embeddings[,3]

# Visualize what headings are called so that you can extract them to form a dataframe
Embeddings(object = data, reduction = "pca")

                                                          PC_1          PC_2         PC_3          PC_4          PC_5          PC_6
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -5.330006e-01 15.7831573486  -4.99912739   4.431528568  12.344715118 -9.942733e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454    -1.523991e+00 13.8032426834  -5.34413099   5.951599121  12.662260056  1.738698e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454    -9.072709e-01 16.6352920532  -5.61871767   4.974217415  13.788206100 -1.106682e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -5.730304e-01 15.6429271698  -5.67932987   5.476871490  13.618507385 -9.325639e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -1.492131e-01 14.8050909042  -4.79194593   4.477123737  12.224570274 -2.831221e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454    -3.660548e-01 15.3729009628  -6.26665592   6.448869705  13.469744682 -4.325381e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454     1.194306e-01 16.8530864716  -6.62010336   7.441605568  14.661459923 -5.750430e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454    -2.065984e+00 15.2632265091  -3.70344591   3.916329861  12.638164520  4.882870e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454    -1.548584e+00 15.0833835602  -5.00290728   3.365054607  12.168499947 -1.200142e+00
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454    -1.805957e+00 14.7095775604  -3.76334620   3.833495140  12.354162216  1.960566e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454    -4.979624e-01 15.6097192764  -6.18945026   6.585252285  13.696672440 -3.300405e-01
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454    -2.043035e+00 10.9974489212  -2.54944658   5.212202549   8.051369667 -7.598011e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454    -8.356372e-01 15.5325307846  -5.96117592   6.481206894  13.347260475 -2.213525e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454    -9.033749e-02 17.6406078339  -7.28156424   8.401016235  15.122898102 -6.395189e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454    -2.914330e+00 -1.6414843798  -3.57089758  -6.497498512   3.247905493 -4.004478e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454    -7.034260e-01 15.8505172729  -5.98290396   6.183456898  13.805769920 -6.636370e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454    -5.494278e-01 16.9386711121  -6.75094748   7.617156029  14.676543236 -4.268895e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454     2.638007e+00 13.2642173767  -3.52007580   3.297828913  11.559779167 -9.327608e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454    -1.063793e+00 14.9242191315  -5.85806179   6.549662113  13.248745918 -7.137018e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454    -9.941164e-01 10.5937366486  -5.71169567   3.632258177  10.855877876 -1.727342e+00
                                                          PC_7          PC_8          PC_9         PC_10         PC_11         PC_12
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -1.5823048353 -0.1460189521 -2.3531355858 -0.4034147859  1.010060e+00 -1.242242e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454     0.8093938828  4.0745525360  2.8720707893 -2.4640698433 -5.469205e-01  5.758511e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454    -1.5789358616  0.0212975442 -1.3608397245  0.3628246188  1.411706e+00 -1.599171e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -1.9826710224 -2.1473784447  0.0107318116 -1.0531636477 -7.399411e-01  8.366549e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -0.2154319286 -1.1156141758  3.6134490967  3.6556010246 -5.766839e-01  7.318997e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454    -1.1579729319 -0.7756161690  2.7647163868 -1.7436710596 -8.506994e-01  8.972740e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454    -1.6920911074 -0.9535923004  3.2410397530 -3.0031275749 -5.564780e-01  4.523649e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454     1.8616019487  6.5816016197 -1.5109727383 -0.0958601087  8.991758e-01 -5.878053e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454    -1.2537747622 -0.6844640374 -2.8660185337  2.1399700642  1.192562e+00 -1.383020e+00
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454     1.7403351068  5.2084922791 -0.3655781746  1.2552425861  5.188878e-01 -2.529318e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454    -1.0582358837 -0.7722041011  3.0721859932 -1.5139043331 -7.109895e-01  6.874685e-01
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454    -0.5476834178  0.2590969801 -0.8020487428  0.7853537798  2.881605e-01  7.694458e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454    -0.4385137856  2.2961506844  2.3085794449 -2.7894232273  7.504693e-01 -1.101765e+00
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454    -1.4882925749  1.3155431747  3.9289636612 -4.6403293610  4.805585e-01 -8.113170e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454    -0.0806917399  0.2646243274  1.1100229025  1.5266330242 -8.217089e-01  1.441019e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454    -0.8599081635  1.2731432915  1.7189826965 -1.8128361702  7.161572e-01 -9.820016e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454    -1.0586160421  1.9431242943  2.9348170757 -3.9571712017  4.592456e-01 -8.290518e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -1.3473163843 -1.0577111244 -2.0048480034  1.2461444139 -8.738574e-01  1.420979e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454    -0.0843934864  2.8753423691  2.6546254158 -3.1262457371  1.063335e-01 -3.022336e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454    -0.3586787581  1.4112807512  4.0291538239 -2.4531519413 -8.304539e-01  4.661430e-01
                                                         PC_13         PC_14         PC_15         PC_16         PC_17         PC_18
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454     7.294953e-01  2.473359e-01 -7.270732e-02 -2.521090e-01 -4.654368e-01 -2.780792e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454    -6.563130e-01 -6.097929e-01  1.682930e-01 -2.196707e-01  3.552442e-01  2.475258e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454     3.170278e-01  6.124064e-01 -3.278669e-01 -3.733237e-01  6.355231e-02  9.698861e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -3.379585e-01 -3.340137e-01  1.280244e-01  2.564772e-01  6.875494e-02 -1.878919e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454     9.090763e-01 -5.115703e-02  2.939613e-01  6.625721e-01 -4.279705e-01 -1.755224e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454    -7.160092e-01 -7.426192e-01  2.459083e-01 -1.848288e-01  3.796791e-01  5.888232e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454    -6.608389e-01 -5.686569e-01  1.608559e-01 -3.444324e-01  3.996090e-01  5.915011e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454     7.845730e-01  8.550652e-01 -2.445948e-01  5.784501e-01 -6.660061e-01 -1.514170e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454     3.351901e-01  4.786427e-01 -2.260917e-01 -2.824205e-01  6.519974e-02  6.015361e-02
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454     8.090588e-01  6.481906e-01 -5.978066e-02  6.044359e-01 -6.449153e-01 -1.525724e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454    -6.048650e-01 -6.486464e-01  2.501285e-01 -2.381237e-01  3.467272e-01  8.273399e-02
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454    -4.757321e-01  3.721789e-01 -2.520603e-01 -1.379211e+00 -7.259366e-01  1.447789e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454    -3.044279e-01  7.736880e-02 -2.484042e-01 -4.632261e-01  4.170283e-01  3.402945e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454    -6.636350e-01 -2.856323e-01 -9.093310e-02 -8.761009e-01  7.093430e-01  2.251630e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454    -2.682567e+00  4.656917e+00 -2.625844e+00  3.152568e+00  2.479281e+00 -3.161562e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454    -1.485458e-01  9.704217e-02 -1.959073e-01 -4.035964e-01  2.870651e-01  2.240526e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454    -5.323181e-01 -1.454448e-01 -1.211627e-01 -6.329373e-01  5.164869e-01  2.341870e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -7.442645e-01  1.214221e+00 -1.654689e+00 -2.010369e+00 -5.546962e-01  1.628046e+00
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454    -5.057213e-01 -3.069415e-01 -4.943693e-02 -4.428465e-01  4.655954e-01  3.297435e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454    -1.474357e+00  8.051838e-01 -6.313702e-01  6.749286e-01  1.101827e+00 -8.776268e-01
                                                         PC_19         PC_20         PC_21         PC_22         PC_23         PC_24
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -3.361792e-02  1.447751e-01  1.208883e-01 -2.611945e-01 -7.184841e-02  7.438824e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454     1.709947e-01  1.347327e-01  4.618974e-02 -1.756053e-01 -1.926692e-01  3.196495e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454     9.360180e-02 -6.518707e-02 -2.673428e-01 -3.187393e-01  9.336621e-02  1.378541e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -1.850931e-01 -5.897813e-02  9.548500e-02  1.927239e-01 -6.593563e-02 -1.193439e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -2.875430e-02 -4.964193e-02  1.782237e-02  1.758711e-01  1.329684e-01 -1.548852e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454    -3.683019e-02 -1.236001e-02  9.205112e-02 -4.512759e-02 -1.551574e-01  1.775038e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454     5.506851e-03  4.424182e-02 -5.439821e-04 -1.358151e-01 -1.185811e-01  3.017917e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454    -1.949050e-01 -1.975647e-02  5.480625e-02  3.971961e-01  3.422881e-01 -5.079930e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454     9.768248e-02 -7.000690e-03 -2.165902e-01 -2.868348e-01  5.089015e-02  9.245434e-02
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454    -5.477002e-02 -5.246362e-02 -6.877988e-04  2.764813e-01  2.532178e-01 -4.490738e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454     4.132291e-02  3.455275e-02  3.084295e-02 -1.738249e-01 -5.805039e-02  3.410251e-01
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454    -1.427839e+00 -1.418182e+00  8.011951e-02 -5.454413e-01  6.865828e-01  4.194458e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454     2.898117e-01  2.011478e-01  1.234212e-03 -1.679437e-01 -4.029469e-01  7.488415e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454     2.666845e-01  7.388177e-02 -1.422659e-01 -6.168332e-01 -1.795759e-01  6.165286e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454     2.402028e+00  5.101840e-01 -5.245062e+00 -4.026399e+00 -4.074595e-01  3.473213e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454     2.337724e-01  1.168464e-01 -6.824884e-02 -1.661357e-01 -2.706376e-01  5.776265e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454     2.044680e-01  2.041140e-02 -1.424628e-01 -4.758938e-01 -2.169030e-01  4.994684e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -2.269364e+00 -2.504901e+00  4.193557e+00 -1.198555e+00 -7.958659e-01 -9.608542e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454     2.418210e-01  1.693951e-01  6.386673e-02 -2.473812e-01 -3.810302e-01  5.966483e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454     5.792484e-01  8.776578e-01 -5.925104e-01 -5.504613e-01 -1.880634e-01  1.186856e+00
                                                         PC_25         PC_26         PC_27         PC_28         PC_29         PC_30
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -3.977298e-01  3.918801e-01 -1.564477e-01 -1.307487e-01 -5.061774e-03 -1.912944e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454     8.510780e-02 -5.265933e-03 -3.134925e-04  1.580443e-01 -4.489630e-02  6.675449e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454     1.192606e-01  5.232215e-02  4.326988e-02 -5.141614e-03  1.449553e-02 -6.223439e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -9.078123e-02  1.396326e-01  3.425571e-02  9.323268e-03 -8.659253e-02 -1.982041e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -3.929860e-02 -2.008294e-01 -2.273200e-01 -2.275601e-01  2.155262e-01  4.303970e-03
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454     5.949186e-02 -2.185955e-02  8.798843e-02  1.338937e-01 -1.398457e-01  1.637987e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454     7.151665e-03 -2.824771e-03  2.423345e-03  1.413002e-01 -1.733714e-01  1.577226e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454    -8.129880e-02  3.419898e-01 -1.093046e-01 -5.941041e-01  3.041223e-01 -1.312581e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454     1.768360e-01  5.848901e-03 -7.766823e-02  1.253037e-04 -1.404598e-01 -6.942572e-02
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454    -6.737050e-02  1.229401e-01 -1.349613e-01 -4.096593e-01  2.963899e-01 -1.458177e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454     8.214568e-02 -3.866295e-02 -1.231640e-02  2.106691e-01 -1.340678e-01  1.600724e-01
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454     4.413525e-01 -2.363106e+00  1.869014e+00  2.721277e+00  4.044595e-01 -3.122877e+00
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454     1.774153e-01 -2.495440e-01  7.436457e-02  3.042828e-01 -3.529809e-01  1.352746e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454     2.887139e-01 -9.830377e-02  1.876618e-02  2.383744e-01 -2.483719e-01  1.343798e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454     2.910516e+00  1.102564e+00 -1.725626e+00  1.483574e+00 -2.135145e+00 -3.559217e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454     1.618010e-01 -2.171500e-01  2.331492e-02  2.023569e-01 -2.544110e-01  1.096027e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454     1.489263e-01 -8.740588e-02 -1.153311e-02  1.832761e-01 -2.104063e-01  7.532500e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454     2.721646e+00  2.827084e+00  8.362729e-01 -8.361251e-02 -5.288748e-01 -9.289427e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454     8.702165e-02 -1.425820e-01  5.902813e-02  2.709337e-01 -2.524801e-01  1.303372e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454     8.782906e-01 -4.602059e-02 -6.118370e-01  4.619919e-01 -4.129651e-01 -8.820707e-01
                                                         PC_31         PC_32         PC_33         PC_34         PC_35         PC_36
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454     7.178337e-02  3.468694e-01 -7.633416e-03  2.547109e-01 -1.607669e-01 -7.588624e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454     1.078478e-01  6.567536e-02 -7.564043e-02 -2.536303e-03 -5.242444e-02  5.642208e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454    -4.433376e-02 -1.821055e-02 -1.274771e-02 -2.635548e-02  1.740603e-02 -7.014634e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454     4.249171e-03 -8.908787e-02 -4.482022e-02  1.245225e-01 -5.678309e-02  1.047219e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454     8.471525e-02  1.038909e-01  2.913630e-01 -1.514539e-02  1.846113e-02  2.000261e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454    -4.943119e-02  1.029804e-01  4.040855e-02 -5.410886e-02 -1.500213e-02  3.431663e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454    -8.906568e-02  4.923093e-02 -8.093462e-02 -1.622516e-01 -8.921239e-02  3.743656e-03
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454     3.415898e-01 -6.192387e-02  2.863931e-01  4.880216e-01 -1.245923e-01  6.956635e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454    -1.635870e-01 -3.102930e-02 -1.014364e-02 -1.447027e-01  4.717002e-02 -4.857706e-02
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454     3.359815e-01  1.729917e-03  2.394908e-01  2.381256e-01 -3.874166e-02  1.233448e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454    -6.318924e-02  3.269618e-02  1.766223e-03 -1.607636e-01 -2.637605e-03 -2.969957e-02
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454    -3.020128e+00 -1.314013e+00 -6.959263e-02  4.193771e-01  3.568034e+00 -6.496847e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454    -1.641549e-01  1.447559e-01 -9.986452e-02 -1.626322e-01  3.822305e-02 -6.053107e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454    -2.412210e-01  9.583429e-02 -1.934880e-01 -3.834872e-01 -4.252978e-02 -9.706888e-02
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454     1.421064e+00 -2.868943e+00 -4.081529e+00  8.878630e+00 -1.958373e+00 -3.553493e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454    -1.385888e-01  8.313509e-02 -7.316940e-02 -1.311843e-01 -7.035925e-03 -7.067559e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454    -2.068422e-01  6.200801e-02 -3.333154e-02 -2.097748e-01 -7.080219e-03 -1.563248e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -4.385982e-01  3.242601e-01  1.186218e+00  1.276476e+00  1.347943e+00  6.258379e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454    -1.014097e-01  1.664639e-01 -1.123762e-01 -1.541795e-01  4.660483e-02 -6.295755e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454     6.077069e-01 -8.327677e-01 -1.265472e+00  2.488956e+00 -5.756503e-01 -7.218584e-01
                                                         PC_37         PC_38         PC_39         PC_40         PC_41         PC_42
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -1.689664e-02 -1.972325e-01 -1.062634e-01 -2.176611e-01  9.353866e-02 -4.498015e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454    -1.305201e-02  1.118238e-01  6.956173e-02  7.286144e-02 -4.204969e-02 -1.045563e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454     7.166870e-02 -1.011380e-02  1.810843e-02 -2.243755e-01  4.089223e-02 -9.096775e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -7.499694e-02 -5.142114e-02  3.235722e-02  2.580668e-02  1.054653e-02 -1.762611e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -2.205022e-01 -2.029201e-01 -2.154740e-02 -9.118601e-02  1.999458e-01 -2.467799e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454     1.521888e-02  5.857362e-02  1.163281e-01  5.914750e-02 -9.708164e-03 -1.818074e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454     7.991549e-02  6.939331e-02  4.535251e-02  6.201423e-02 -7.918986e-02 -2.174866e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454    -1.575505e-01 -1.614078e-01 -1.657474e-01  1.251460e-01  1.143411e-01  2.847401e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454     3.849934e-02 -6.665115e-02  2.792093e-02 -4.471807e-02 -8.694945e-02 -3.725541e-02
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454    -1.841896e-01 -1.848598e-01 -8.723698e-02  1.321984e-01  6.782246e-02  2.036870e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454     5.889064e-02  1.073408e-01  8.057758e-02  2.929789e-03 -1.008963e-01 -2.270185e-01
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454     1.613025e+00  7.014663e+00 -9.202596e+00  2.949777e+00 -9.664982e+00  9.640714e+00
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454     8.275391e-02  1.372719e-01  4.264903e-02 -9.388711e-02 -2.176736e-02 -2.024437e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454     1.958274e-01  2.096978e-01 -1.173926e-01  3.414507e-02 -8.306266e-02 -2.904435e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454     2.831384e+00  2.893290e+00 -4.650798e-01 -2.104443e+00  1.311148e+00 -3.450706e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454     9.760256e-02  1.212680e-01  4.582188e-02 -6.033961e-02 -6.719584e-02 -1.621660e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454     1.486137e-01  1.128113e-01  4.917265e-02  8.891997e-03 -4.511118e-02 -2.889845e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -1.765312e-01  7.473551e-01  8.582608e-01 -9.052635e-01 -1.210818e-02 -1.176940e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454     7.303432e-02  1.488307e-01  5.052372e-02 -6.745781e-03 -7.876925e-02 -1.919898e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454     5.032282e-01  9.111993e-01 -4.038033e-01 -7.601706e-01  7.583009e-02 -9.282212e-01
                                                         PC_43         PC_44         PC_45         PC_46         PC_47         PC_48
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -1.846406e-01  1.547253e-01  2.297880e-01 -1.943068e-02 -2.792307e-01 -5.937706e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454    -5.772941e-02 -4.995238e-02  2.229435e-01 -1.269658e-02 -1.112332e-01 -6.078280e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454     5.303998e-02  1.857944e-02  1.150127e-01 -2.350276e-02 -9.493674e-02 -1.608055e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454    -1.631701e-01  3.260004e-02  6.515490e-02 -1.316489e-03 -8.854987e-02 -9.888419e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -1.484205e-01 -1.083003e-01 -3.071407e-01  1.440026e-01 -7.384166e-02  2.190104e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454    -7.780109e-02 -4.700582e-02  3.027983e-01  2.758679e-02 -4.392372e-02  1.004395e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454     1.618318e-01 -1.340839e-01  3.122295e-01  1.081941e-03 -5.903462e-02 -6.968346e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454    -5.004801e-01  9.769507e-02 -8.189958e-01 -1.185895e-01 -1.391004e-02  1.722746e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454     4.028716e-02 -5.255036e-02  6.954242e-02  5.470438e-02  3.286935e-02 -1.326982e-01
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454    -2.477535e-01  6.182216e-02 -4.947552e-01 -6.486649e-02 -1.633124e-01  4.576390e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454    -1.846112e-02 -5.164485e-02  2.833284e-01  5.407869e-02 -8.365037e-02 -9.971757e-02
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454    -2.116442e+00 -3.688505e+00 -3.158458e+00  1.017042e+00 -3.195291e-01 -1.171085e+00
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454     6.187797e-02 -1.471764e-01  3.143212e-01  7.808868e-02 -9.036778e-02 -1.025820e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454     1.241250e-01 -1.913793e-01  6.055672e-01  1.857070e-01  1.007440e-01 -1.963456e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454    -3.849472e+00 -8.260834e+00  8.411742e+00 -5.707857e+00  1.167274e+01  7.241825e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454    -3.108031e-02 -1.104532e-01  2.975808e-01  9.492937e-02 -3.928870e-02 -1.307017e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454     2.075522e-02 -3.319533e-02  4.071069e-01  1.319857e-01  8.243781e-02 -1.094437e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -7.953045e-01  1.077939e+00 -5.254813e-01 -9.795715e-01 -7.058629e-01  7.108837e-01
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454     5.553305e-02 -7.521629e-02  3.535968e-01  8.646125e-02 -1.496579e-01 -8.141145e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454    -1.188960e+00 -2.053133e+00  2.597486e+00 -8.228294e-01  3.381056e+00  2.191938e+00
                                                         PC_49         PC_50
2637-SM-1-GCCAAT_S1:AAAACCTCCCACTCCTCx-DMS454    -7.585093e-02 -8.387165e-02
2637-SM-1-GCCAAT_S1:AAAACCTCCACGAAACGx-DMS454     6.375924e-02 -8.397289e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTATAGTCGCAx-DMS454     6.255343e-02 -1.203182e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCTTTACCCTx-DMS454     2.751223e-02  9.196898e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTACACAAGGCx-DMS454    -2.439860e-01 -2.369405e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACCCTAACCx-DMS454     7.985099e-02 -1.775235e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCCCGCAACTx-DMS454     9.817056e-02 -1.786353e-01
2637-SM-1-GCCAAT_S1:AAAACCTCCACCACGCTx-DMS454    -4.951902e-01  3.509490e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAATACTCTTx-DMS454    -4.071366e-02 -1.129273e-02
2637-SM-1-GCCAAT_S1:AAAAGTCGGAATTCCCAx-DMS454    -3.041050e-01  5.165340e-02
2637-SM-1-GCCAAT_S1:AAAAGCCTACATCGCAGx-DMS454     9.131834e-02 -1.183159e-01
2637-SM-1-GCCAAT_S1:AAAACTCGATTGTTTACx-DMS454     3.167262e-01 -4.674257e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAGCAAAGCCx-DMS454     1.243121e-01 -1.805104e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATAAATAGGx-DMS454     1.355283e-01 -4.413607e-01
2637-SM-1-GCCAAT_S1:AAAACTCGACTACGAGCx-DMS454    -2.490258e+00  2.252123e+00
2637-SM-1-GCCAAT_S1:AAAACCTCCAGCAGAACx-DMS454     1.126873e-01 -1.382249e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTAATGGATTAx-DMS454     6.485464e-02 -3.004072e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTACGTATTTCx-DMS454    -4.851449e-01  4.000901e-02
2637-SM-1-GCCAAT_S1:AAAACTCGAAAACAGGGx-DMS454     1.172132e-01 -1.553439e-01
2637-SM-1-GCCAAT_S1:AAAAGCCTATCCAGTCCx-DMS454    -7.841925e-01  4.542918e-01
 [ reached getOption("max.print") -- omitted 16088 rows ]

plot.data <- FetchData(object = data, vars = c("PC_1", "PC_2", "PC_3", "cline"))
plot.data$label <- paste(rownames(plot.data))
plot_ly(data = plot.data, 
        x = ~PC_1, y = ~PC_2, z = ~PC_3, 
        color = ~cline, 
        colors = c("lightseagreen",
                   "green",
                   "red",
                   "orange1",
                   "royalblue1",
                   "lightcyan3",
                   "peachpuff3",
                   "darkorchid1",
                   "turquoise",
                   "darkmagenta"),
        type = "scatter3d", 
        mode = "markers", 
        marker = list(size = 5, width=2), # controls size of points
        text=~label, #This is that extra column we made earlier for which we will use for cell ID
        hoverinfo="text") #When you visualize your plotly object, hovering your mouse pointer over a point shows cell names

VizDimLoadings(data, dims = 1:3, reduction = "pca")

Save files

data <- ProjectDim(data, reduction = 'pca')

PC_ 1 
Positive:  NR2F1, COL2A1, ZIC2, NELL1, FABP7, DLX6-AS1, MAP1B, CALB1, RIPPLY3, CDKN1A 
       IGFBP5, RALYL, SPOCK1, MYCN, SMOC1, PCDH9, POSTN, S100A6, MS4A8, FAM60A 
Negative:  CAV1, RPS27A, SGK1, MT-CO2, CTGF, SERPINH1, LTBP4, LPHN2, MT-ND1, CLDN6 
       EIF5A, MT-ND4, MTATP6P1, MYC, HMGA1, CAV2, HMGA2, MT-CO1, MID1, RPL37A 
PC_ 2 
Positive:  ASCL1, DSP, SCNN1A, CALCA, KRT8, UCP2, RAB3B, SEC11C, MPP1-1, HMGB3-1 
       TMEM176B, PCSK2, UGDH, SST, TFF3, AGT, SCN3A, OR51E2, ABHD2, BCAP31-1 
Negative:  TMSB4X, PTMS, TUBB, ENC1, CKB, FSCN1, MAP1B, TCF4, CSNK1E, MYC 
       TMSB4XP8, NUCKS1, ACTG1, DCX, EEF1A2, TUBA1A, RPL14, SGK1, RPS28, CAV1 
PC_ 3 
Positive:  CAV1, SGK1, CTGF, SERPINH1, NR2F1, CLDN6, LPHN2, CAV2, MYL12B, CRIP2 
       PXDN, S100A11, MYL12A, PTRF, COL2A1, ANXA1, MMP10, RDH10, MID1, ALDH1A1 
Negative:  INSM1, BASP1, HEPACAM2, FAM91A1, MEST, CDKN2C, DLK1, MYC, PCSK2, NHLH2 
       RPL37, DPYSL2, COTL1, RAI14, NKX2-2, TSPAN13, SRD5A1, CDH12, SLFN11, ELN 
PC_ 4 
Positive:  CAV1, SGK1, PIM1, ENC1, CTGF, PCSK2, MAP2, BASP1, KIF1A, HEPACAM2 
       LPHN2, CLDN6, SERPINH1, FAM91A1, VMP1, TENM3, ARID1A, FNIP2, EIF5A, MID1 
Negative:  SOX11, SNTB1, RPS28, MYC, RPL18, TPT1, NPW, RPL5, RPL21, ASS1 
       SUSD2, BEX1, RPL23, RTN1, RPS24, LDHB, RPL24, HMGB1, RPL19, RPLP1 
PC_ 5 
Positive:  DSP, PCSK2, MPP1-1, TDRD1, TMEM176A, G6PD-1, AMBP, MIAT, POU4F2, CLDN18 
       GUCY2C, PLCG2, ONECUT2, MIB2, AKR1B10, OR51E2, MTSS1, CLU, UGDH, CMIP 
Negative:  PCSK1, SST, BCAT1, CCND2, FAM178B, PEG10, SPINK1, RAPGEF5, FAM91A1, MEST 
       CYP1A1, TAGLN2, UCP2, DLK1, EGFL7, EPCAM, KLK11, NHLH2, RAI14, GADD45A 

Archetypal Analysis

We will now apply the archetype analysis to the single cell data. Since we’ve reduced the PCs to 10, we have gotten rid of a lot of the noise and still captured a large proportion of the variance.

Choosing number of archetypes

Fit to k = 2 to 8 to find the best number of archetypes. We will look at the variance explained by each archetype as well as the t-ratios. To choose a final number to move forward with the analysis, we will run a randomization test to get a p-value for each number of archetypes (t-ratio test).

library(ParetoTI)

Loading required package: data.table
data.table 1.14.2 using 1 threads (see ?getDTthreads).  Latest news: r-datatable.com
**********
This installation of data.table has not detected OpenMP support. It should still work but in single-threaded mode.
This is a Mac. Please read https://mac.r-project.org/openmp/. Please engage with Apple and ask them for support. Check r-datatable.com for updates, and our Mac instructions here: https://github.com/Rdatatable/data.table/wiki/Installation. After several years of many reports of installation problems on Mac, it's time to gingerly point out that there have been no similar problems on Windows or Linux.
**********

Attaching package: ‘data.table’

The following object is masked from ‘package:SummarizedExperiment’:

    shift

The following object is masked from ‘package:GenomicRanges’:

    shift

The following object is masked from ‘package:IRanges’:

    shift

The following objects are masked from ‘package:S4Vectors’:

    first, second

Loading required package: lpSolve
Registered S3 methods overwritten by 'dbplyr':
  method         from
  print.tbl_lazy     
  print.tbl_sql      
Warning: replacing previous import ‘ggplot2::last_plot’ by ‘plotly::last_plot’ when loading ‘ParetoTI’

library(cowplot)
library(ggplot2)
library(RColorBrewer)
library(reshape2)

Attaching package: ‘reshape2’

The following objects are masked from ‘package:data.table’:

    dcast, melt

library(factoextra)

Welcome! Want to learn more? See two factoextra-related books at https://goo.gl/ve3WBa

library(ggfortify)
library(cluster)
# install.packages("data.table")
##################################
# load(file="../../out/cell-lines/ParetoTI/X_magic_for_ParetoTI.Robj")

x <- data[["RNA"]]@scale.data

x_pca <- read.csv('../../out/cell-lines/ParetoTI/pca_embedding.csv',header = TRUE,row.names = 1)
x_pca <- t(x_pca)

x_pca <- x_pca[1:11,] #keep only top 11 PCs
loadings<- read.csv('../../out/cell-lines/ParetoTI/pca_feature_loadings_projected.csv', header = TRUE, row.names = 1)
loadings <- as.matrix(loadings)

arc_ks_8 = k_fit_pch(x_pca[1:8,], ks = 2:8, check_installed = T,
                   bootstrap = T, bootstrap_N = 200, maxiter = 1000,
                   bootstrap_type = "m", seed = 2543, 
                   volume_ratio = "t_ratio", # set to "none" if too slow
                   delta=0, conv_crit = 1e-04, order_type = "align",
                   sample_prop = 0.75)

# Show variance explained by a polytope with each k (cumulative)
plot_arc_var(arc_ks_8, type = "varexpl", point_size = 2, line_size = 1.5) + theme_bw()

plot_arc_var(arc_ks_8, type = "res_varexpl", point_size = 2, line_size = 1.5) + theme_bw()

plot_arc_var(arc_ks_8, type = "total_var", point_size = 2, line_size = 1.5) +
  theme_bw() +
  ylab("Mean variance in position of vertices")

plot_arc_var(arc_ks_8, type = "t_ratio", point_size = 2, line_size = 1.5) + theme_bw()

Warning: Removed 1 row(s) containing missing values (geom_path).
Warning: Removed 1 rows containing missing values (geom_point).

arc_ks = k_fit_pch(x_pca, ks = 2:8, check_installed = T,
                   bootstrap = T, bootstrap_N = 200, maxiter = 1000,
                   bootstrap_type = "m", seed = 2543, 
                   volume_ratio = "t_ratio", # set to "none" if too slow
                   delta=0, conv_crit = 1e-04, order_type = "align",
                   sample_prop = 0.75)

Warning: could not honor request to load desired versions of Python; '/Users/smgroves/Documents/anaconda3/envs/reticulate_PCHA/bin/python3.7' was loaded instead (see reticulate::py_config() for more information)

# Show variance explained by a polytope with each k (cumulative)
plot_arc_var(arc_ks, type = "varexpl", point_size = 2, line_size = 1.5) + theme_bw()

plot_arc_var(arc_ks, type = "res_varexpl", point_size = 2, line_size = 1.5) + theme_bw()

plot_arc_var(arc_ks, type = "total_var", point_size = 2, line_size = 1.5) +
  theme_bw() +
  ylab("Mean variance in position of vertices")

plot_arc_var(arc_ks, type = "t_ratio", point_size = 2, line_size = 1.5) + theme_bw()

Warning: Removed 1 row(s) containing missing values (geom_path).
Warning: Removed 1 rows containing missing values (geom_point).

The plots above show a few things. 1. The 7th and 8th archetypes do not add much variance explained ( k = 6 explains over 80% of the variance, whereas 7 and 8 explain ~ 8% combined). 2. When we look at the variance explained on top of k-1 model, the 6th archetype clearly explains a large proportion of the variance (more than 5, actually) 3. The 7th and 8th archetypes have higher variance in position of the vertex, suggesting these are less robust than the first 6 which ave close to 0. 4. The t-ratio shows an increase for 6 archetypes from 5, suggesting it fits the geometry of the data better than 5 archetypes. We will follow this up with t-ratio tests to confirm.

T-ratio tests

i = 6
arc <- fit_pch(x_pca, noc = i, delta = 0, conv_crit = 1e-04, maxiter = 500)

start = Sys.time()
pch_rand = randomise_fit_pch(x_pca, arc_data = arc,
                             n_rand = 1000,
                             replace = FALSE, bootstrap_N = NA,
                             volume_ratio = "t_ratio",
                             maxiter = 500, delta = 0, conv_crit = 1e-4,
                             type = "m", clust_options = list(cores = 3))
# use type m to run on a single machine or cloud
# type = "m", clust_options = list(cores = 3))
# use clustermq (type cmq) to run as jobs on a computing cluster (higher parallelisation)
# type = "cmq", clust_options = list(njobs = 10)) 


pdf(sprintf('../../figures/ParetoTI/%s_t-ratio_test.pdf', i))
plot.r_pch_fit(pch_rand, type = c("t_ratio"), nudge_y = 5)
dev.off()
  # This analysis took:
print(Sys.time() - start)



#    k  var_name   var_obs p_value
# 1: 3   varexpl 0.4309729   0.001
# 2: 3   t_ratio 0.5023466   0.001
# 3: 3 total_var        NA     NaN
# ---                             
#   k  var_name   var_obs p_value
# 1: 4   varexpl 0.5727041   0.001
# 2: 4   t_ratio 0.2134559   0.003
# 3: 4 total_var        NA     NaN
# ---
#    k  var_name   var_obs p_value
# 1: 5   varexpl 0.6979477   0.001
# 2: 5   t_ratio 0.1492318   0.001
# 3: 5 total_var        NA     NaN
# ---
#    k  var_name   var_obs p_value
# 1: 6   varexpl 0.8313096   0.001
# 2: 6   t_ratio 0.2436056   0.001
# 3: 6 total_var        NA     NaN

Fitting archetypes using PCHA

We’ll start by fitting 6 archetypes and finding the enriched genes and gene sets for each.

Idents(object = data) <- data@meta.data$cline #add cell line labels

cols <- c(brewer.pal(9, "Set1"),'gray')

plot_arc(arc_data = arc, data = x_pca,
                   which_dimensions = 1:2,colors = cols,
                  data_lab = as.character(Idents(data))) + theme_bw()
p_pca = plot_arc(arc_data = arc, data = x_pca, 
                 which_dimensions = 1:3, line_size = 1.5,
                 colors = cols,
                  data_lab = as.character(Idents(data)),
                 text_size = 60, data_size = 2) 
plotly::layout(p_pca, title = "Archetypes for Top 10 PCs")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

htmlwidgets::saveWidget(p_pca, "../../figures/ParetoTI/robust_archetypes.html")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

p_pca = plot_arc(arc_data = arc_ave, data = x_pca, 
                 which_dimensions = 1:3, line_size = 1.5,
                 data_lab = as.numeric(data[['RNA']]@scale.data['NEUROD2',]),
                 text_size = 60, data_size = 3) 
plotly::layout(p_pca, title = "Expression of NEUROD2 in PCA")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

htmlwidgets::saveWidget(p_pca, "../../figures/ParetoTI/NEUROD2.HTML")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

p_pca = plot_arc(arc_data = arc_ave, data = x_pca, 
                 which_dimensions = 1:3, line_size = 1.5,
                 data_lab = as.numeric(data[['RNA']]@data['YAP1',]),
                 text_size = 60, data_size = 3) 
plotly::layout(p_pca, title = "Expression of YAP1 in PCA")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

htmlwidgets::saveWidget(p_pca, "../../figures/ParetoTI/YAP1.html")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

p_pca = plot_arc(arc_data = arc_ave, data = x_pca, 
                 which_dimensions = 1:3, line_size = 1.5,
                 data_lab = as.numeric(data[['RNA']]@data['POU2F3',]),
                 text_size = 60, data_size = 3) 
plotly::layout(p_pca, title = "Expression of POU2F3 in PCA")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

htmlwidgets::saveWidget(p_pca, "../../figures/ParetoTI/POU2F3.html")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

p_pca = plot_arc(arc_data = arc_ave, data = x_pca, 
                 which_dimensions = 1:3, line_size = 1.5,
                 data_lab = as.numeric(data[['RNA']]@data['ASCL1',]),
                 text_size = 60, data_size = 3) 
plotly::layout(p_pca, title = "Expression of ASCL1 in PCA")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

htmlwidgets::saveWidget(p_pca, "../../figures/ParetoTI/ASCL1.html")

No trace type specified:
  Based on info supplied, a 'scatter3d' trace seems appropriate.
  Read more about this trace type -> https://plotly.com/r/reference/#scatter3d
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...
A marker object has been specified, but markers is not in the mode
Adding markers to the mode...

Archetypes: 1. N (H524) 2. A/N (CORL279) 3. A2 (DMS53) 4. Y (H841) 5. A (H69) 6. A2 (DMS454) # Determine enriched genes and gene sets to define archetypes

This will be especially interesting for comparing archetypes 4 and 5, since it is pretty clear that the shape of the data is not clearly defined without an archetype at 5, but not many cells actually lie close to it. We use the ParetoTI package to evaulate gene sets enriched in each location.

labs = get_top_decreasing(summary_genes = enriched_genes, summary_sets = enriched_sets,
                          cutoff_genes = 0.05,cutoff_sets = 0.05,
                          cutoff_metric = "wilcoxon_p_val", 
                          p.adjust.method = "fdr", 
                          order_by = "mean_diff", order_decreasing = T)

 --  archetype_1

SNTB1, SUSD2, MYC, CRABP1
HMGB1, JPH4, SOX11, GABBR2
NPW, RPL38, HMOX1, RPL21

heart_growth
T_helper_cell_differentiation
interleukin_13_production 

 --  archetype_2

BASP1, FAM91A1, MEST, DLK1
HEPACAM2, NHLH2, ISL1, INSM1
RAI14, MEX3D, CDKN2C, PEG10

regulation_of_interleukin_13_production
regulation_of_cardiac_muscle_tissue_growth
negative_regulation_of_CD4_positive__alpha_beta_T_cell_differentiation 

 --  archetype_3

SST, PCSK1, UCP2, SPINK1
FAM178B, CCND2, SEC11C, SCNN1A
EGFL7, CYP1A1, TFF3, EPCAM

determination_of_bilateral_symmetry
atrial_septum_morphogenesis
signal_transduction_involved_in_G2_DNA_damage_checkpoint 

 --  archetype_4

CAV1, SGK1, CTGF, CLDN6
SERPINH1, LPHN2, CAV2, ENC1
PXDN, PTRF, MID1, FNIP2

metanephric_epithelium_development
exosomal_secretion
atrial_septum_morphogenesis 

 --  archetype_5

NR2F1, COL2A1, ZIC2, NELL1
FABP7, CALB1, CDKN1A, MAP1B
DLX6-AS1, RALYL, SPOCK1, PCDH9

regulation_of_glomerular_mesangial_cell_proliferation
glomerulus_vasculature_development
atrial_septum_development 

 --  archetype_6

PCSK2, DSP, MPP1-1, ASCL1
TDRD1, MIAT, UGDH, POU4F2
G6PD-1, CLDN18, OR51E2, GUCY2C

extracellular_exosome_biogenesis
cerebral_cortex_GABAergic_interneuron_differentiation
positive_regulation_of_CD4_positive__alpha_beta_T_cell_activation 

Save enrichments to be compared to bulk archetypes

The format we need the enrichment file in is a csv with the columns: archetype #,Feature Name,P value (Mann-Whitney),Median Difference,Mean Difference,Significant after Benjamini-Hochberg correction?,Is first bin maximal?

enriched_sets$`Feature Name` <- mutate_all(enriched_sets$`Feature Name`, funs=toupper)

Error in UseMethod("tbl_vars") : 
  no applicable method for 'tbl_vars' applied to an object of class "character"

What direction are various mutation vectors facing? Can we uncover the driver mutations like Uri Alon has done?

What is the difference between CORL279 and N+ cell lines? They seem to have the same shape, but CORL279 is translated (shifted) above N cell lines. Can we characterize that vector in PCA space? In archetype space?

We find the vector associated withthis shift by taking the difference of the averages of each cell line in PCA space (or archetype space). This should result in a vector pointing from one average to the other. We can then reconstruct these vectors in gene space to determine which genes are playing a major role in the difference between the two populations.

session_info. = devtools::session_info()
session_info.